Diagnostic Yield of Modified Bleach-Assisted Ziehl-Neelsen Microscopy in Cervical Tuberculous Lymphadenitis: A Prospective Paired Diagnostic Accuracy Study

Background: Direct Ziehl-Neelsen (ZN) microscopy is accessible but has limited sensitivity in paucibacillary tuberculous lymphadenitis.We evaluated whether a modified bleach concentration method increases the detection of microbiologically confirmed acid-fast bacilli (AFB) when used as an adjunct to direct ZN microscopy.Materials and methods: This prospective, single-center, paired diagnostic accuracy study included 124 consecutive patients with clinically suspected cervical tuberculous lymphadenitis.For each participant, two separate microscopy slides were prepared from aliquots of the same lymph node aspirate: a direct ZN smear without bleach pretreatment and a second smear after modified bleach pretreatment and concentration, followed by the same ZN staining protocol.A composite microbiological reference standard based on cartridge-based nucleic acid amplification testing and/or mycobacterial culture was available for 116 participants.The primary outcome was paired sensitivity among reference-positive participants.Secondary outcomes included incremental confirmed yield; rescue, defined as modified-bleach positivity among specimens negative by direct ZN microscopy; specificity and other accuracy measures among participants with an available composite microbiological reference result; the cytomorphological distribution of discordant results; and exploratory scenario-based laboratory cost.Accuracy estimates used exact 95% confidence intervals, and paired differences were tested with the two-sided exact McNemar test.Results: Direct ZN microscopy detected AFB in 21/124 specimens (16.9%), whereas modified bleach-assisted ZN microscopy detected AFB in 26/124 specimens (21.0%).Among 38 reference-positive participants, sensitivity was 55.3% (95% CI, 38.3-71.4) for direct ZN and 68.4% (95% CI, 51.3-82.5)for modified bleachassisted ZN microscopy, a numerical difference of 13.2 percentage points.The prespecified two-sided exact McNemar comparison did not reach the conventional threshold for statistical significance (p = 0.0625).No false-positive result was observed among 78 reference-negative participants, corresponding to an observed specificity of 100% (95% CI, 95.4-100.0%)for each method.Modified bleach-assisted ZN microscopy detected 5/103 ZN-negative specimens (4.9%; 95% CI, 1.6-11.0),and all five had a positive composite microbiological reference result.It rescued 5/17 reference-positive cases missed by ZN (29.4%; 95% CI, 10.3-56.0).Four discordant cases occurred in necrotizing suppurative lymphadenitis, and one occurred in necrotizing granulomatous lymphadenitis.An exploratory scenario using locally reconstructed process-cost inputs estimated a cost of approximately Rs 256 per additional confirmed detection.Conclusions: Modified bleach-assisted ZN microscopy increased microbiologically confirmed AFB yield and the point estimate of sensitivity, without an observed loss of specificity.Because inference was driven by only five discordant pairs and reference testing was incomplete in eight participants, the method should be considered a low-cost adjunct to direct ZN microscopy rather than a replacement for WHO-recommended nucleic acid amplification testing.